The recent financial crisis has exposed the global banking system to a series of adverse shocks. The crisis started with a substantial fall in the value of mortgage-related securities which inflicted heavy losses on the banking sector. In some countries, a liquidity crisis has developed as uninsured deposits were suddenly withdrawn, which intermittently led to bank failures.
Governments have made extensive efforts to provide assistance to the banking sector, as the continued weakness of this sector was perceived to have negative effects on the real economy. The present paper examines these effects and provides new evidence on the sources, magnitude and duration of the real effects of banking shocks.
The paper estimates the impact of variations in bank profits, bank capital and bank reserves on GDP growth and other real economic variables. The sample comprises a panel of 30 OECD countries observed annually for 9 to 24 years starting in 1979. While the sample includes several episodes of bank crises, about 95% of the observations refer to non-crisis years. Hence, the paper documents the real effects of banking shocks that prevail regularly in non-crisis periods. Nevertheless, the findings of this paper are also relevant for crisis episodes, where the shocks hitting the banking sector are particularly large.
The main finding is that variations in bank profits have a significant impact on GDP growth lasting approximately two years. Other things equal, one percentage point decline in bank ROA (return on assets) is expected to reduce the following year's GDP growth by 0.3 percentage points. The effect is stronger for economic activities and industrial sectors that rely more heavily on external finance, and is more pronounced in countries with a large banking sector. Bank reserves (cash) also exhibit some impact on real economic activity, though to a lesser extent than ROA. Surprisingly, variations in bank capital do not show any significant impact on the real economy. This result is consistent with recent findings of Giannetti and Simonov (2009) who showed that recapitalization of Japanese banks during the 1990s did not have significant effects on the real economic activity of their clients.